Video Blog

2019 Combined Sections Meeting, Platform Presentation

Predicting Discharge Locomotor Function for Non-Ambulatory Individuals following Stroke who Participate in Focused Stepping Practice during Inpatient Rehabilitation

Chris Henderson PT, PhD, Jenni Moore PT, DHS, T. George Hornby PT, PhD

Awarded Best Platform by the Stroke Special Interest Group

Purpose/Hypothesis: Neuromuscular impairments underlying locomotor deficits are common post-stroke and rehabilitation interventions early following injury are directed towards mitigating these impairments. However, converging evidence suggests that the amount of high-intensity task specific practice provided may be an important, modifiable attribute of rehabilitation that may magnify or accelerate locomotor improvements. Previous work identified that changes in walking function with inpatient rehabilitation were positively associated with the amount of task-specific practice provided, although recovery varied across patients.  The purpose of this study was to identify specific predictors of locomotor recovery at discharge from inpatient rehabilitation, from both initial evaluation and early intensity rehabilitation to ameliorate walking deficits.  

Participants: 255 non-ambulatory individuals (i.e., requiring at least minimal assistance to ambulate) < 6 months post-stroke in inpatient rehabilitation consented to participate in this study.

Materials/Methods: Demographics (age, comorbidities, stroke type and distribution) and clinical scores (paretic lower extremity strength, Berg Balance Score [BBS], Six Minute Walk Test distance and level of assistance, and 10 Meter Walk Test) during the 1st, 2nd, and discharge week of inpatient rehabilitation were extracted from patients’ medical records.  Receiver operating characteristic (ROC) curves were generated for individual demographic and clinical measures to identify optimal cut-off scores for discriminating ambulatory function at discharge. The area under the curve was used in turn to qualify the utility of the cut-off. Additionally, separate forward stepwise logistic regression analyses were performed with data available during the 1st or 2nd week of admission to predict the likelihood of independent ambulation (contact guard or above) at discharge from inpatient rehabilitation.

Results: ROC curves indicate all clinical variables had at least ‘acceptable’ ability to discriminate discharge walking function, while all demographic measures were ‘poor’. Specifically, 1st week BBS scores ≥ 5 best discriminated between individuals that were and were not able to walk independently at discharge from inpatient rehabilitation. Across all clinical measures, areas under the curve were greater at 2nd week of assessment when compared with 1st week assessment. Logistic regression analyses using either admission BBS and paretic lower extremity strength or 2nd week BBS alone were best able to predict independent ambulation at discharge.

Conclusions: Prediction of ambulatory function at discharge from inpatient rehabilitation is feasible for individuals post-stroke participating in a stepping focused program, with patients with very poor initial BBS scores achieving substantial gains in locomotor function with focused stepping practice.  Clinical measures at initial evaluation provide good discrimination of walking function at discharge, with prediction enhanced with 2nd week outcomes indicating responsiveness to rehabilitation. Balance and strength were found to be key predictors of discharge walking function, agreeing with previous work; however, application of current findings may only be appropriate with attempts at similar interventions during inpatient physical therapy.

Clinical Relevance: Early prediction of walking function at discharge from inpatient rehabilitation is important to facilitate appropriate discharge planning. 


1.         Moore, J. L., Roth, E. J., Killian, C. & Hornby, T. G. Locomotor training improves daily stepping activity and gait efficiency in individuals poststroke who have reached a ‘plateau’ in recovery. Stroke 41, 129–135 (2010).

2.         Van Middendorp, J. J. et al. A clinical prediction rule for ambulation outcomes after traumatic spinal cord injury: A longitudinal cohort study. Lancet 377, 1004–1010 (2011).

3.         Veerbeek, J. M., Van Wegen, E. E. H., Harmeling Van Der Wel, B. C. & Kwakkel, G. Is accurate prediction of gait in nonambulatory stroke patients possible within 72 hours poststroke? The EPOS study. Neurorehabil. Neural Repair 25, 268–274 (2011).

4.         Bland, M. D. et al. Prediction of discharge walking ability from initial assessment in a stroke inpatient rehabilitation facility population. Arch. Phys. Med. Rehabil. 93, 1441–1447 (2012).

5.         Holleran, C. L., Rodriguez, K. S., Echauz, A., Leech, K. A. & Hornby, T. G. Potential Contributions of Training Intensity on Locomotor Performance in Individuals with Chronic Stroke. J. Neurol. Phys. Ther. 39, 95–102 (2015).

6.         Hornby, T. G. et al. Variable Intensive Early Walking Poststroke ( VIEWS ): A Randomized Controlled Trial. (2015). doi:10.1177/1545968315604396

7.         Hornby, T. G. et al. Feasibility of Focused Stepping Practice during Inpatient Rehabilitation Poststroke and Potential Contributions to Mobility Outcomes. Neurorehabil. Neural Repair 29, 923–932 (2015).

8.         Ivey, F. M., Stookey, A. D., Hafer-Macko, C. E., Ryan, A. S. & Macko, R. F. Higher Treadmill Training Intensity to Address Functional Aerobic Impairment after Stroke. J. Stroke Cerebrovasc. Dis. 24, 2539–2546 (2015).

9.         Boyne, P. et al. High-Intensity Interval Training and Moderate-Intensity Continuous Training in Ambulatory Chronic Stroke: Feasibility Study. Phys. Ther. 96, 1533–1544 (2016).

10.       Munari, D. et al. High-intensity treadmill training improves gait ability, VO2peak and cost of walking in stroke survivors: preliminary results of a pilot randomized controlled trial. Eur. J. Phys. Rehabil. Med. (2016). doi:10.23736/S1973-9087.16.04224-6